3 Binomial and Pc '--cn Probability Distril 3.1 Binomial Experiment Definition: An experiment which satisfies the following conditions is called a binomial experiment. De experiment chad consist of Finite number of small experiment Ccatiect trials). Cwe say we have n number of small experiments) 2) Each smallay experimend Cérials) can have only two possivlé outcomes, ank they ave game for all trtals-Cwe denote two ert outcomes by sucess) and failuve (2). Dov the erials should be Mhependent A tre probability of obtaming a success BD or Sailuve €) 75 same from trial to trial. Cwe denote the success probability as P) Definition: Given a binomial experiment consisting of n trials with success probability p, we may define the random variable X=number of successes obtained: ‘Then, X is known as the binomial random variable and denoted by cv BC1>P) Example: The same coin is tossed successively and independently 3 times. Dn=d = nis Finite- D Each trial dw can have heal or tail 2) AV 3 trials ave independent frerm each other det probability of obtaining a head ov tai) ave Same for al trials: co this $5 o binomial experiment Note that S = {SSS, SSF, SFS,SFF,FSS,FSF,FFS,FFF}andX € {0,1,2,3}. whet X= Number of heals obtarned. Then XavpC%e-) 1Activity: Determine wiscther the following given: experiments are binomial experiments or not 1) Asking 1000 people if they vote for candidate Ms. A. C Suppos € that there are no any preferences im voting for ms. A ¥ thts 75 a binomial experiment- Herve = n=l00e 2) Asking 200 people if they watch XTVnews.( Aesume there are ve any Prefevences jy) wotching XTV news, £ Bimewral experim ont 3) Rolling a die until a 6 appears. * Not a bivemial experiment - pee fecause nis unknown ank n con be nfinjke- 4) Rolling a die to see if a5 appears. C Bolling only once) ¥ Binomial experiment and two possible outcomes ial Here nel, : Lh not havig a five having oo 5 an 5) Asking 20 people how old they are. x Not a binemtal experiment: Because we hove more thon twe orbeomes: (The age of vondomly selected person cam be any’) 6) Drawing 5 cards from a deck for a poker hand. H Wet a binomial experiment * Because the 5 trials ave nor imbependent ; 2bC*5" oP) =(ayP*G-ey [Pwr (Lys ea eral ) * PC of a binomial randoe earable XV denoted by Fio:9 p) and given by “ (ert ph O12 wat Oronp) IP mila {f otherwise Bach of six randomly selected cola drinkers is given a glass containing Scola ant one lisse ace identical in appearances except for a onde on the boltomn to identify the cola. Suppose there is actually no tendeney among cola drinkers to prefer one coln to the other. Find the probability that (a) exactly & drinkers prefer Seoln. (bj at least 8 drinkers prefer Seobe Here n-6 , Lek X= Number of divin drinkers prefer Scola - then p = oS a pox =3) ~ bC 356,09) oe as Geet = CA. os ous ies D plx>s) = Pcx= o +PCx= 0 + PCX™ HDepex-) = )—Cpex-c) + pCx=1) + PCX=2) )-[bC05 6,02) + BCI; be A) + BC as 6 2D] = use the formule ont find the vatue- Proposition. For X ~ Bin(n, p), olf is denoted by P(X <2) = Blrsn.p) = SH DAZ PC KSA) = BOTS MP) = LCOSmP)EbCS Sor) ¥bCr5n5f) Example. Suppose that 20% of all copies of « particular textbook fail a certain binding strength test. Find the probability that (a) at most 8 fail the tet (b) exactly & fail the test, of a sample with 15 copies. het X= Number of copies fail the test: ou df then p= Roy = 0-2. we, o) pC x <8) = BC85 1850-2) 2-994, CFrom the table) - DpCx-8) = bC#5 1807) 9 Gr ° K pC K-68) -= bCH3,0-2) = BC#5, 0%) - B C7315 02) 15-8 .2%(-0-2) it pe tt 3 = 0°999— 9-996 by ad7567 F oak od A ie a = 00037 <————_+34 Binomnis’ bilities » ee ee ee) " sh) SOK cs am 9) 800m 1 noo “sui? om we ey“ 2 sot 816 8K a8 nt oe e890 os eas ar vale ‘008 em 4 oo ORT RM 6s M7 an oe oka ‘ tom hea mes 8 EN Ph ee ae ee fom 1k 9) lima ss as n86 Yo aK 999) nk Vom 19 0) 10) (oor Van Loon 90) Lou me ORD boo 10 Ln eK) Lem Lom tom Vogt Le LOB) 1K LL La ee ee tha) nek 810 em fist 902 km Net 013 a0, 0K 01M an, 0m aa Onna tes Ja SIT ao Proposition. For X ~ Bin(n,p), then © B(X) = np © VIX) = apt —p) = mpg © oy = YT (where q = Example. If 73% of all purchases at a certain store are tade with « credit card, find the average aumber of purchases made with a credit card, among ten randomly selected purchases. What is the standard deviation? Let X= Number of purchaes marl with a credit coma then P= Ts). =0°78 4» N10 Them, » E(x) = “P = oxor7s = 7°54 # VR) = np C1-P) = jOX o-PSRX OAS a ba = sd = By, 2 rere zs 19364 2 oi yf -P)jin with pareaucier ACA >) Defiration. A random Varivble X is id to have a Poiswon distred if the pmfef X is on PX =) plgay SE 0.1.2 The value of A is the average number of wbcevations that goour i ecified time oF region Example, Let X denote the suber of creatures of « particular vps caplurod in a trap durias given time period. Suppose that X fs a Poisson distribution with A 4.5. sa on average ( will contain 4.5 creatures. Find the probability that a trap contains exactly five ereatures X= wamber of creatures D= 45 Ain PCxes)= Ca nt ee 5S ee 4-3 5s! a wold ye © The Poisson distribution expresses the probability of « given number of events occurring in a xed interval of time and/or space if these events occur with a known avernge rate and independently of the time since the Inst event Pinf for the Poisson distribution; + Applications of the Poisson distribution inelude; 0.40 © The number of telephone calls arriving in a system within a given time interval. (Tclocommuniestion) © Photons arriving at a telescope. (Astronomy) © Thommmber of mutations on « strand of DNA per unit longth. (Biology) 00] © Customers arriving at.a counter or call center. (Busi- 9 99] ness) © Cars arriving at a traffic light. (Engineering) & is the number of occurrences. Note that this is a discrote distribution.550 and np<5, thev the correspond ny - fag distribution com be approxi wort ed by he a persion distribukion: Then bonsnyp) & p(n3a), heve A=MP tion = Bi Proposition. Suppo that in the binwioal pul bapnp), we ki h—y o@ and p> Vin suck 4 4s that np a= $= 0 Then (snp) apg) reece + Rute of thumb: ‘This epproximution ean safely be applied i Example. Ifa publisher of nontechnical books tukes grent pains to ensure that its books are free of y of any given p 1 at least one sich error is te page, what is the probability that one of its 400:puge cypographical errors, so that the probs (05 and errors are independent fom p novels will contain exactly one page with evr Experiment =3 checking errors im poges of books. Let X = Mumbex of pages containing & feasl on€ evvor Kiewe note that the two outcomes ave page do not conta 2 con ab estone ere ang ervey eve wWenno, - > e-7Pe and ee = zs « yoeke-005 = % : ce gape 5 ee potsten() > with a=nP Bim Caveye-2 c ree Cisa)- ea Se tee (Hee bC15 40% ar : ; 2 i550 90,0 008)* Proposition. If X has Poisson distribution with parameter A, {X ~ Po(A)), then LX) = V(X) Example, Let X denote the number of creatures of « particular type captured in a trap during a siven time period. Suppose that X has w Poisson distribution with A= 4.5, so on average traps will contain 4.5 ercatures. Find the expected number of creatures trapped anid the varianee of the sumber trapped. KEK) = AH=4-5P RVKY HA RaSy Suppose pulscs arrive at 8 counter at an average rate of six por minute. Find the probability that in a 30 sec interval at least one pulse is received. xeA> 6 =3 Let x=Number of pulses arrive # pCX et) = IH PC x=0)6 a eee aaa tae eo? as ois y